ok - to your original question. In the Veterinary diagnostic testing world we would take into account the consequences of a wrong answer that was worse than the target medical condition. This usually drove us to improve either the sensitivity or specificity of the test. This usually involved the risk of a false positive - would the unneeded treatment cause harm to the patient? Was that harm worse than the harm of the untreated disease? For example a false positive for heartworm would result in near fatal treatment. Of course in diagnostic testing veterinarians (and human doctors) will almost always use other coincident or secondary means to confirm a diagnosis before taking drastic treatment actions…Note that this involves both knowing the occurrence rate and implementing specific protocols.
It could also involve a false negative: for example a pre-surgery blood test for clotting ability that came back OK when the patient actually had little to no clotting ability would be in real danger of bleeding out on the operating table. A false negative on a parvo test is fatal not only to the patient but to any other patients and even the breeder’s business or the practice’s business. This drives very high accuracy in these tests. And constant monitoring and release testing to ensure that this almost never happened. Difficult to implement a protocol for these tests so the emphasis was on the product accuracy.
And as always, stated label claims about effectiveness helps. Know the occurrence rate, do not rely on guessing.
Stay away from wild hypotheticals (meteor strikes and the zombie apocalypse) and deal only with the product and risks at hand.
The restraint example is a good example. Stating the limitations of the device and the mitigating protocol is essential. AND as a manufacturer you should make every effort to ensure your product and its usage conditions are mitigated. While we can never eliminate all harms we can reduce most of them to a negligible level. (Now I know that some medical treatments are almost as bad as the disease - chemo for example. It is intended to harm the cancer but also harms other bodily tissue. To be clear I am not talking about that. This really is a balancing act that must be weighed. Is the treatment worse than the disease? And what other choices do people have?)
It could also involve a false negative: for example a pre-surgery blood test for clotting ability that came back OK when the patient actually had little to no clotting ability would be in real danger of bleeding out on the operating table. A false negative on a parvo test is fatal not only to the patient but to any other patients and even the breeder’s business or the practice’s business. This drives very high accuracy in these tests. And constant monitoring and release testing to ensure that this almost never happened. Difficult to implement a protocol for these tests so the emphasis was on the product accuracy.
And as always, stated label claims about effectiveness helps. Know the occurrence rate, do not rely on guessing.
Stay away from wild hypotheticals (meteor strikes and the zombie apocalypse) and deal only with the product and risks at hand.
The restraint example is a good example. Stating the limitations of the device and the mitigating protocol is essential. AND as a manufacturer you should make every effort to ensure your product and its usage conditions are mitigated. While we can never eliminate all harms we can reduce most of them to a negligible level. (Now I know that some medical treatments are almost as bad as the disease - chemo for example. It is intended to harm the cancer but also harms other bodily tissue. To be clear I am not talking about that. This really is a balancing act that must be weighed. Is the treatment worse than the disease? And what other choices do people have?)
